A new approach, based on control charts, is presented for the task of recognition of events and scenarios in video image sequences. For each image in the sequence, low level image processing and feature extraction steps result in feature descriptors for objects of interest detected in the images. Control charts analysis is then explored to classify the nature of the activity depicted by the temporal changes in these features over the image sequence. Scenario recognition with higher accuracy is achieved using this simple approach.
|Original language||English (US)|
|Number of pages||7|
|Journal||Turkish Journal of Electrical Engineering and Computer Sciences|
|State||Published - Dec 1 2005|
ASJC Scopus subject areas
- Computer Science(all)
- Electrical and Electronic Engineering